The endocannabinoid system (ECS) is comprised of two cannabinoid receptor subtypes (CB1 and CB2), their endogenous ligands (i.e., the endocannabinoids anandamide and 2-arachidonoyl glycerol), and enzymes for ligand biosynthesis and degradation (e.g., monoacylglycerol lipase, fatty acid amide hydrolase). The ECS plays a prominent role in the regulation of a variety of physiological functions, including control of food intake and energy metabolism, emotional behavior, pain, cell division, and inflammation. CB1 receptors are widely expressed in numerous peripheral organs and tissues, including thyroid gland, adrenal gland, reproductive organs, bone, adipose tissue, liver, muscle, pancreas, kidney, and gastrointestinal tract. CB1 receptors have also been identified in brain, including cortex, hippocampus, amygdala, pituitary and hypothalamus. CB2 receptors are largely localized in immune and blood cells, but have more recently been identified in brain [for reviews see Endocrine Reviews 2006, 27, 73; Int J Obesity 2006, 30, S30; J Clin Endocrin Metab 2006, 91, 3171; Int J Obesity 2009, 33, 947].
Many disease states, including inflammatory and metabolic diseases and certain cancers are associated with overactivity of the ECS system. This is characterized by increased ECS tone in peripheral tissues including adipose, liver, muscle, and pancreas. Elevated ECS tone is reflected by increased tissue expression of CB1 receptors as well as elevated tissue levels of the main endogenous cannabinoids anandamide and/or 2-arachidonoyl glycerol. Preventing and/or reversing overactivity of the ECS system has proven to be a useful approach toward the treatment of inflammatory and metabolic diseases and certain cancers [Mol Pharmacol 2003, 63, 908; J Clin Invest 2008; 118:3160; Diabetes 2010, 59, 926; Cancer Res 2011, 71, 7471; Cell Metab 2010; 11:273; J Biol Chem 2008; 283:33021; Int J Obesity 2007, 31, 692].
The plant-derived cannabinoid agonist Δ9-tetrahydrocannabinol (Δ9-THC), the main psychoactive component of marijuana, binds to both CB1 and CB2 receptors. Δ9-THC is widely reported to increase appetite and food intake (hyperphagia) in humans and in animals. This hyperphagic effect is largely blocked by pretreatment with CB1 antagonists and inverse agonists (e.g., rimonabant, taranabant, otenabant, ibipinabant), drugs that effectively block the CB1 receptor, strongly supporting the belief that CB1 receptor activation mediates the hyperphagic effect of Δ9-THC, [Endocrine Reviews 2006, 27, 73].
In humans, rimonabant and taranabant produce a clinically meaningful weight loss in obese patients. Obese patients also experience improvements in diabetic and cardiometabolic risk factors associated with obesity, including an increase in the level of high density lipoprotein cholesterol (HDL), and decreases in triglycerides, glucose, and hemoglobin A1c (HbA1c, a marker of cumulative exposure to glucose) levels. Rimonabant also produces reductions in abdominal fat deposits, which are a known risk factor for diabetes and heart disease [Science 2006, 311, 323]. Taken together, these improvements in adiposity and cardiometabolic risk factors produce an overall decrease in the prevalence of the metabolic syndrome [Lancet 2005, 365, 1389 and NEJM 2005, 353, 2121].
In patients with type 2 diabetes not currently treated with other anti-diabetic medications, rimonabant was shown to significantly improve blood sugar control and weight, as well as other risk factors such as HDL and triglycerides, when compared to placebo. After six months of treatment, HbA1c levels were significantly lowered by 0.8% from a baseline value of 7.9 as compared to a reduction of 0.3% in the placebo group [(Diabetes Care 2008, 31, 2169; Lancet 2006, 368(9548), 1660-7]. Rimonabant also improved glycemic control and cardiometabolic risk factors in type 2 diabetic patients receiving insulin [Diabetes Care 2010, 33, 605]. These results are consistent with preclinical studies that demonstrate improved glycemic and lipid control in diabetic and dyslipidemic mice, rats, and dogs (Pharmacology Biochemistry & Behavior, 2006, 84, 353; American Journal of Physiology, 2003, 284, R345; American Diabetes Association Annual Meeting, 2007; Abstract Number 0372-OR).
The beneficial effects of rimonabant on diabetic and cardiometabolic risk factors such as high blood pressure, insulin resistance, and elevated triglycerides cannot be explained by diet-related weight loss alone. For example, in patients receiving 20 mg of rimonabant, only approximately 50% of the beneficial effects on triglycerides, fasting insulin, and insulin resistance can be accounted for by weight loss secondary to reduced food intake. These results suggest a direct pharmacological effect of CB1 antagonists on glucose and lipid metabolism, in addition to indirect effects on metabolism secondary to hypophagia-mediated weight loss [Science 2006, 311, 323 and JAMA 2006, 311, 323]. Taken together, these results suggest that CB1 antagonists might be effective in the treatment of diabetes, dyslipidemias (e.g., high triglycerides and LDL, low HDL), cardiovascular disorders (e.g., atherosclerosis, hypertension), and hepatic disorders (e.g., fatty liver diseases, non-alcoholic steatohepatitis, cirrhosis, liver cancers), even in patients that are not clinically overweight or obese.
The CB1 receptor is overexpressed in alveolar rhabdomyosarcoma (ARMS), a pediatric soft tissue tumor derived from skeletal muscle, and upregulation of CB1 is a diagnostic marker for ARMS [Cancer Res 2004, 64, 5539]. CB1 overexpression is essential for tumor cell proliferation and metastasis, and the CB1 inverse agonist AM251 abrogates both cell invasion and lung metastasis in vivo [Cancer Res 2011, 71, 7471]. The CB1 inverse agonist rimonabant has also been demonstrated to inhibit human breast and prostate cancer proliferation [Mol Pharmacol 2006, 70, 1298; Cancer Res 2005, 65, 1635], and to inhibit human colon cancer cell growth and reduce the formation of precancerous lesions in the mouse colon [Int J Cancer 2009, 125, 996].
The CB1 receptor is one of the most abundant and widely distributed G protein-coupled receptors in the mammalian brain. It is now known that the appetite-suppressant properties of CB1 antagonists can be mediated through either a direct action with CB1 receptors in brain regions associated with hunger and satiety (e.g., hypothalamus, mesolimbic regions), or a direct action with CB1 receptors in peripheral tissues (e.g., adipose tissue, kidney) [J. Clin Invest 2010, 120: 2953; Obesity 2011, 19: 1325] However, CB1 receptors are far more broadly distributed in brain (e.g., neocortex, hippocampus, thalamus, cerebellum, and pituitary), and while interacting with targeted CB1 receptors in hypothalamus and mesolimbic regions to suppress appetite, CB1 antagonists have equal access to non-targeted CB1 receptors that have little if any role in appetite control. Binding to non-targeted receptors can often lead to unwanted side effects of CNS drugs [Endocrine Reviews 2006, 27: 73]. The CB1 blockers rimonabant and taranabant produce psychiatric and neurological side effects. These include depressed mood, anxiety, irritability, insomnia, dizziness, headache, seizures, and suicidality.
These side effects are dose-related and appear pronounced at the most efficacious weight-reducing doses of rimonabant and taranabant (JAMA 2006, 311, 323; Cell Metabolism 2008, 7, 68). The occurrence of therapeutic efficacy (appetite suppression) and side effects over the same dose range strongly suggest that both effects are mediated through concurrent blockade of CB1 receptors in both ‘targeted’ and ‘non-targeted’ brain regions. Brain-penetrant CB1 blockers do not selectively target CB1 receptors in efficacy brain regions, while ignoring CB1 receptors in side effect brain regions.
The beneficial effects of the CB1 antagonist rimonabant on body weight, adiposity, and diabetic and cardiometabolic risk factors such as high blood pressure, insulin resistance and blood lipids cannot be explained by weight loss derived from CNS-mediated appetite suppression alone [JAMA 2006, 311, 323]. At least 50% of the benefit is likely derived from an interaction with CB1 receptors in peripheral tissues known to play an active role in metabolism. These include adipose tissue, liver, muscle, pancreas, kidney, reproductive tissues, and gastrointestinal tract.
In view of the above, it is highly desirable to find effective and highly selective CB1 receptor blockers with limited or no CNS adverse side effects, including mood disorders. Particularly, it is desirable to find compounds that preferentially target CB1 receptors in peripheral tissues (e.g., adipose tissue, liver, muscle, pancreas, reproductive tissues and gastrointestinal tract), while sparing CB1 receptors in brain. In this way, peripherally-mediated beneficial effects of CB1 blockers can be maintained, whereas CNS side effects will be reduced or eliminated. This should provide a novel opportunity to develop safer alternatives to highly brain penetrant CB1 blockers for the prevention or treatment of obesity, diabetes, dyslipidemia, cardiovascular disorders, hepatic disorders, and/or certain cancers.